Home
Page

Texting
Data

Word
Cloud

We created a word cloud of the words we text to each other most often below:

“Baby”, “Babe”, “Love”, and “Lol” are a few of our most frequently used words!

Our texts over time

Sent Texts Quantity

Our first winter together, texting ~600 times a week was the norm. Since Neil moved in, our texting slowed down quite a bit.

^{Hover over the chart and milestones for more information about our milestones and texting. Or click on the chart legend to show/hide any milestones or trends.}

Average Character Count Per Text

There seems to be a slight downward trend in the length of the texts that we send each other - however, certain events seem to spike longer texts towards one another. The week after we got engaged, Neil’s texting character count increased dramatically!

^{Hover over the chart and milestones for more information about our milestones and texting. Or click on the chart legend to show/hide any milestones or trends.}

Total
Texts

Below are two bar charts showing the total number of texts we have sent to each other and the average number of text characters in each of our text messages to each other.

Neil doesn’t always respond! Karina has sent over 1,000 more texts to Neil since they have known each other.

^{Click on the chart legend to show/hide any bars.}

But when Neil responds, he has more to say! Neil’s sent texts are 5 characters longer, on average.

^{Click on the chart legend to show/hide any bars.}

Our Favorite Words to Text

Overall

In the chart below, we selected words that we use frequently in texts to each other, or we feel are special to us.

Look below to see that Karina says “Baby” a lot, and Neil “Laughs Out Loud” a lot.

^{Click on the chart legend to show/hide any bars. (Note: Ben is the name of Karina’s son and Laika is the name of Neil’s dog!).}

Over Time

Take a look at our trends in the selected words we text each other below!

For better interpretability, we have highlighted certain sections of this graph to tell a few stories. Check out the tabs below to see some interesting trends we found in or texting!

“Love
you”

Right after we decided to make it an official relationship, “Love you” made its way into our regular texting. We were quick to say we love each other… And we love** that!**

^{Click on the chart legend to show/hide any milestones or trends.}

“Baby” vs “Babe”

At the beginning of their relationship, Neil rarely used the word “Baby” but overtime he seems to mimic the way Karina used the word. An increase in Neil texting the word “Baby” can be seen after we move in together.

^{Click on the chart legend to show/hide any milestones or trends.}

Furthermore, if you look at the trend of the word “Babe” among each of us, Neil has used “Babe” much less in his current texts than with Karina, but Karina rarely used that word to start with!

^{Click on the chart legend to show/hide any milestones or trends.}

An increase in “Ben”

Notice how Ben (Karina’s son) becomes a topic of conversation much more frequently after he was introduced to Neil.

^{Click on the chart legend to show/hide any milestones or trends.}

^{Note: One of Neil’s best friends is also named Ben, so some of this data may be inflated.}

Neil Likes to “Lol”

After we made our relationship official, Neil started to “Laugh Out Loud” much more frequently. Karina “Laughed Out Loud” at a pretty low frequency throughout their relationship.

^{Click on the chart legend to show/hide any milestones or trends.}

The Fall of “Good Morning”

From December through February, we were saying “Good morning” to each other nearly everyday. Since the end of February, the number of times we say “Good morning” to each other via text is nearly zero. Although we started “officially” living with each other in August, we have been seeing each other every morning for quite a while. We get to wish each other a good morning in person everyday, now! The only time we text each other “Good morning” or “Good night” now is when one of us are away.

^{Click on the chart legend to show/hide any milestones or trends.}

All
Words

^{Click on the chart legend to show/hide any milestones or trends.}

Combined
Data

^{Click on the chart legend to show/hide any milestones or trends.}

Travel
Data

A collage of us together in some of our favorite trips!

A Map of the places we have traveled together!

^{Hover over the markers to get information and see when we last visited each place!}

Snowboarding
Data

Snowboarding is one of our favorite hobbies and we get data from our Mt. Hood Meadows annual ski passes. Thanks to Mt. Hood Meadows for or snowboard data!

In the first tab, we display the runs at our local ski resort, Mt. Hood Meadows ski resort. In the next tab we wanted to show everyone how often we like to snowboard, using the data we get from our annual ski passes.

Take a look at the charts of our snowboarding data in the tabs below!

Mt. Hood Meadows Ski Resort

Mt. Hood Meadows Ski Resort Runs on Google Maps.

^{Hover over the marker to see where we got engaged!}

Mt. Hood Trail Map

Take a look at how awesome our local ski resort is. We love Mt. Hood Meadows!

Chairlift
Data

Over Time

Each bar below represents the amount we used each ski lift at Mt. Hood Meadows ski resort on each date we snowboarded.

By Vertical Feet Gained

^{Hover over each dot to see how much vertical feet we gained on each date we snowboarded at Mt. Hood Meadows ski resort.}

By number of Runs Completed

Hover over each dot to see how many runs we took on each date.

^{Hover over each dot to see how many times we took the ski lift on each date we snowboarded at Mt. Hood Meadows ski resort.}

Overall

Each box below represents the amount we used the ski lifts at Mt. Hood Meadows ski resort.

In vertical feet gained

^{Hover over each box to see how much vertical feet we have gained on each chairlift at Mt. Hood Meadows ski resort!}

By number of runs

^{Hover over each box to see how many times we have taken each chairlift at Mt. Hood Meadows ski resort!}

Financial
Data

Our Overall Spending Habits

Each square below corresponds to the percentage disposable income (i.e. necessary expenses like mortgage, utilities, car payments, student loans, etc. are not included) we spend on selected categories. We seem especially interested in spending our disposable income on bars/restaurants and Amazon products!

^{Hover over each box to see what percentage we spend on each category}

Our Individual Spending Habits

Karina’s Spending

Karina spends the most percentage of her disposable income on Amazon purchases and other merchandise!

^{Hover over each box to see what percentage Karina spends on each category.}

Neil’s Spending

Neil spends the majority of his disposable income on bars & Restaurants. Makes sense, he loves to eat and drink!

^{Hover over each box to see what percentage Neil spends on each category.}

Market
Comparisons

As seen in the tabs below, our relationship milestones seem to have a huge impact of the S&P 500 and Dow Jones Market increases. Since we first met, both the S&P 500 and Dow Jones the markets have risen dramatically!

We like to contribute each markets’ success to our milestones!

Market Trends

The S&P 500 Market & our milestones

^{Hover over the chart and milestones for more information about our milestones and the market price. Or click on the chart legend to show/hide any trends.}

The Dow Jones Market & our Milestones

^{Hover over the chart and milestones for more information about our milestones and the market price. Or click on the chart legend to show/hide any trends.}

Texting Data & Market
Correlates

For fun, we decided to see if our texting data is predictive of the S&P 500 and Dow Jones market prices. We share a passion for working with quantitative data and this was a fun way for us to show off that passion. Karina works regularly with data models in her job in asset management and Neil used to teach statistics to undergraduate/graduate college students.

According to the models in the table below, when correlated to the absolute market prices of the S&P 500 and Dow Jones markets (controlling for time), our texting is predictive of the change in the absolute market price. The relationship is positive and statistically significant.

However, when compared to the daily difference between the start/close of each market price (controlling for time), our models were no longer predictive.

^{Note: These models are completely for fun and should not be take seriously.}

Main Effect Model Tables and visuals

Market Name	Unstandardized Beta	Degrees of Freedom	F-value	p-value	R	R-squared
S&P 500 close price (Absolute)	0.47	(2, 408)	2003.30	< 0.01	-0.26	0.91
S&P 500 close price (Daily Difference)	0.02	(2, 408)	0.11	0.9	0.02	< 0.01
Dow Jones close price (Absolute)	3.49	(2, 408)	750.28	< 0.01	-0.23	0.79
Dow Jones close price (Daily Difference)	-0.09	(2, 408)	0.04	0.96	-0.01	< 0.01
Predictor variable = Daily quanitity of texts sent between Neil and Karina Outcome variable = Corresponding market closing price (Absolute closing value or Daily difference closing value) All models controlled for time

Model Scatterplots & Interpretations

S&P 500 (Absolute)

Interpretation: This model is statistically significant: F(2, 408) = 2003.30, p <.01, R² = 0.91. Suggesting that, for every 1 text we send, the S&P 500 market price is expected to increase 0.47 points, after controlling for time. The beta (slope) for our texts was significant at p < 0.05, indicating that, after controlling for time, our texts have a very very slight influence on the market. The overall model explains 91% of the variance in the market price (Mostly explained by time, but it’s fun to think our texts influence it a little bit!).

S&P 500 (Daily Difference)

Interpretation: This model is not statistically significant: F(2,408) = 0.11, p >.05, R² < 0.01. Our texting is not predictive of the S&P 500 daily difference price.

Dow Jones (Absolute)

Interpretation: This model is statistically significant: F(2, 408) = 750.28, p < .01, R² = 0.79. Suggesting that, for every 1 text we send, the Dow Jones market price is expected to increase 3.49 points, after controlling for time. The beta (slope) for our texts was significant at p < 0.05, indicating that, after controlling for time, our texts have a very very slight influence on the market. The overall model explains 79% of the variance in the market price (Mostly explained by time, but it’s fun to think our texts influence it a little bit!).

Dow Jones (Daily Difference)

This model is not statistically significant: F(2,408) = 0.04, p >.05, R² < 0.01. Our texting is not predictive of the Dow Jones daily difference price.

Interaction Model Table and Visuals

For the more statistically fluent, we also found an interaction between number of texts sent & time on the S&P 500 absolute closing price. As can be seen in the models below, in the beginning of our relationship, our texts were predictive of an increase in the market values. However, the more time that passes, the relationship between our texting and market value turns negative! Therefore, the more time that passes, and the more we text, the worse the absolute market value is expected to be!

Market Name	Unstandardized Beta (texts)	Unstandardized Beta (texts*time)	Degrees of Freedom	F-value	p-value	R	R-squared
S&P 500 close price (Absolute)	1.07	-0.0028	(3, 407)	1347.84	< 0.01	-0.26	0.91
Predictor variable = Daily quanitity of texts sent between Neil and Karina Outcome variable = Corresponding market closing price (Absolute closing value or Daily difference closing value) All models controlled for time and the interaction term between texting and time is included

S&P 500 interaction (Absolute)

Interpretation: This model is statistically significant: F(3, 407) = 1347.84, p <.01, R² = 0.91. Suggesting that, for every 1 text we send, the S&P 500 market price is expected to increase 1.07 points - however, for every day that passes, the beta (slope) is expected to decrease by 0.0028 points. The chart above shows the expected change in slope at 0, 200, 400, 600, & 800 days since we started texting. As can be seen from the chart above, after enough time, the relationship turns from positive to negative!

_{Interested in how these charts were made?: Click here to see our Github repository!}